KR100366143B1 - Method and apparatus for data coding / decoding and coded data recording medium - Google Patents

Abstract

Image data using intra-frame coding and / or predictive coding to receive image data (e.g., video data) and to provide one I-picture and two consecutive P-pictures. Coded image that is generated to generate position information indicative of the position of two P-pictures associated with an I-picture and to record I-pictures, two P-pictures, and positional information on a recording medium An apparatus and a method for recording data on a recording medium are provided. The coded image data is reproduced from the recording medium in a specific reproduction mode by selectively reading the data using the position information.

Description

Method and apparatus for data coding / decoding and coded data recording medium

The present invention relates to a data coding method and apparatus capable of coding video and audio data for specific reproduction, and a data coding method and apparatus for reading video and audio data recorded from an optical disk, a magnetic disk, etc. and reproducing data read in a specific mode Decoding method and apparatus, and also relates to a recording medium recorded in such a manner that the coded data can be reproduced in a specific mode.

Digital video signals to be recorded on a disc in a digital video disc (hereinafter referred to as DVD) system are compressed and coded using MPEG (Motion Picture Coding Experts Group).

14A is a diagram schematically showing an inter-frame prediction structure used in an MPEG system. In this example, a group of pictures (GOP) includes, for example, one frame of an I-picture (intra-frame encoded image), a P-picture (an inter- , And the remaining 10 frames consisting of a B-picture (forward and backward bi-predictive encoded pictures).

An I-picture is a coded picture within a data-compressed frame of one frame, and the P-picture is a compressed inter-frame forward prediction (I-picture or P-picture) with reference to a temporary preceding frame inter-frame forward predictive coded picture, and further wherein the B-picture is a predictively coded bi-directional predictive coded picture with reference to a temporally preceding and following frame,

In particular, as shown by the arrows in the figure, the I-picture I ₀ is coded by an inner frame which is self-processed without referring to any other frame, and the P-picture P ₀ is coded by the I- ₀ ), and the P-picture P ₁ is coded by inter-frame prediction with reference to the P-picture P ₀ . B-pictures B ₀ and B ₁ are coded by interframe prediction with reference to both I-picture I ₀ and P-picture P ₀ , and B-pictures B ₂ and B ₃ are coded by intra- Is coded by inter-frame prediction with reference to both the P-picture (P ₀ ) and the P-picture (P ₁ ). Similarly, a continuous picture is coded by this prediction in the manner indicated by the arrows.

When decoding the predictive coded pictures described, the I-pictures are decoded alone because they are not coded with reference to any other frame. However, since the P-picture is a temporally preceding I-picture and is predictively coded with reference to the P-picture, a preceding I-picture or a preceding P-picture is required to decode a predetermined P-picture. Likewise, to decode a given B-picture, a preceding and subsequent I-picture or P-picture is required because the B-picture is temporarily coded with reference to an I-picture or a P-picture.

For this reason, in order to provide adequate decoding, the position of the picture on the recording medium is changed to the position shown in Figs. 14A to 14B so that the picture for which decoding is required is first decoded.

As shown in the figure, such a position change is B- picture (B- B- ₁ and ₂₎ since the I- to the image (I ₀₎ required to decode the I- picture (I ₀₎ B- picture (B - ₁ and B- ₂ and the position of the P-picture P ₀ is such that the decoding of the B-pictures B ₀ and B ₁ is preceded by the I-picture I ₀ and the P- ₀ ), so that it is changed so as to precede the B-pictures B ₀ and B ₁ . Similarly, the other picture image B- (B ₂ and B ₃₎ Since the P- picture (P ₀ and P ₁₎ is required for decoding P- picture (P ₁₎ the B- image (B ₂ and B ₃ ) and to the preceding, and B- picture (B ₄ and B ₅₎ P- picture (P ₂₎ the B- image (B ₄ and B _5, so the decoding is required for P- picture (P ₁ and P ₂₎ of the ). ≪ / RTI > In the same way, the position change is made such that the P-picture P ₃ precedes the B-pictures B ₆ and B ₇ .

Video data composed of I-picture, P-picture and B-picture arranged in the order of Fig. 14B and other data including audio data and subtitle (caption) data are packetized (multiplexed) Or transmitted over a transport channel. The code quality of each frame in image data is not fixed between images, but depends on the complexity or flatness of each image. Typically, an I-picture appears with more data than a P-picture that appears as more data than a B-picture.

15A to 15C show examples of how data is packetized. In this figure, Fig. 15A shows the multiplexed MPEG2 system flow after packetization, Fig. 15B shows the contents of the video packet in the multiplexed stream, and Fig. 15C shows the MPEG2 video stream in the video layer.

Each image data V, V + 1, V + 2,. . . , Picture header information and picture coding extension information are attached to the preceding position. In the illustrated example, the video stream range from the location identified by D1 to the location identified by D3 of the video layer forms a video packet as long as it has a packet header appended to its preceding location, and from location D3 to D5 of the video layer The video stream range to the identified location forms another video packet with a packet header appended to its preceding location.

This packetized video packet is multiplexed into an audio packet and a subtitle packet thereby forming the MPEG2 system flow shown in FIG. 15A.

Fig. 16 shows the contents of the picture header, and Fig. 17 shows contents of the picture coding extension.

The picture header includes information items such as a single picture_start_code, a temporal_reference (TR) (which is a predetermined serialization serial number), and picture_coding_type (I-, P-, or B-picture).

The picture coding extension includes information items such as a single extension_start code, a single extension_start_code_identifier, picture, top_field_first, progressive_frame, and the like.

With respect to image data, two types of data structures, that is, a frame structure in which one image consists of one frame and a field structure in which one image consists of two fields can coexist. Whether or not the field structure is a frame structure in which the image data is a frame equivalent to the picture or whether the field structure is two fields equivalent to the picture is determined by the following three information items: (1) presence or absence of a GOP header; (2) temporal_reference And 3) picture_structure in the picture coding extension.

18 is a block diagram for explaining an example of a data decoding apparatus applied to execute specific reproduction of data such as slow image reproduction, fast reproduction, reverse reproduction and the like. The optical disc 1 is rotatable by a spindle motor (not shown) at a predetermined rotation rate and the laser beam is projected from the pickup 2 onto a track on the optical disc 1 and recorded on the track MPEG compressed digital data is read. The digital data is processed by the demodulation circuit 3 for demodulating the 8-to-14 modulation (EFM) and supplied to the sector detection circuit 4. [ The output of the pickup 2 is also supplied to a phase locked loop (PLL) circuit 9 where the clock signal is regenerated and supplied to the demodulation circuit 3 and the sector detection circuit 4.

The digital data recorded on the disk 1 includes a sector header and a multiplexed stream recorded as a unit of a fixed length sector together with a sector sync attached to the beginning of each sector. The sector detection circuit 4 detects each sector from the sector sync and the sector address from the sector header, and this information is supplied to the control circuit 6.

The decoded digital data is supplied to an ECC (Error Correction) circuit 33 which performs error detection and correction through the sector detection circuit 4. [ The ECC circuit 33 supplies the error-corrected data to the ring buffer 5 so as to be recorded under the control of the control circuit 6.

The output of the ECC circuit 33 is also supplied to the stream detector 50 to determine the image type in the data stream picture header read from the disc 1 in the specific reproduction mode and supply the picture type information to the control circuit 6 do. In response to this information, the control circuit 6 executes the control operation in such a manner that the data of the two P-pictures continuous with the data of the I-picture in the specific reproduction mode are recorded in the ring buffer 5. [

A focus control circuit (not shown) and a tracking servo circuit 8 are connected to a system controller (not shown) in response to a focus error signal and a tracking error signal obtained from the information read by the pickup 2, respectively Controls the focus and tracking of the pickup 2 under control.

The control circuit 6 designates a write address for writing a corresponding sector in the ring buffer 5 with the write pointer WP according to the sector address of each sector detected by the sector detection circuit 4. [ In addition, in accordance with the code request signal obtained from the video code buffer 10 (Fig. 18B), the control circuit 16 also specifies the read address of the data written to the ring buffer 5 by the read pointer RP. The control circuit 6 reads the data from the position of the read pointer RP and supplies the read data to the demultiplexer 32. [

Since the coded data recorded on the disk 1 is composed of multiplexed video, audio, and subtitle data, the demultiplexer 32 converts the data read from the ring buffer 5 into video data, audio data, And supplies each data to the video decoder 20 (FIG. 18B), the audio decoder (not shown), and the subtitle decoder (not shown). The video decoder 20 stores the video data in the video code buffer 10.

The data stored in the video code buffer 10 is then supplied to an image header detector 34 for detecting the image header. The detected picture header information is also used to identify the temporal reference TR indicating the frame order in the GOP and the picture type (I, P, or B picture) of the video data. The image data selection circuit 35 selects only the I-picture and the P-picture in accordance with the image type information supplied from the image detector 34 in the specific reproduction mode, and outputs the selected image data to the variable length coding To the circuit 11. In the normal reproduction mode, the image data selection circuit 35 is controlled to transfer all the image data to the inverse VLC circuit 11 without arbitrarily pre-selecting it.

The data supplied to the inverse VLC circuit 11 is processed using the inverse VLC and supplied to the inverse quantizer 12. The code request signal is returned to the video code buffer 10 from the inverse VLC circuit so that new data can be delivered in the video code buffer 10.

The inverse VLC circuit 11 also outputs the quantization step size to the inverse quantizer 12 and outputs the motion vector information to the motion compensator 15. [ The quantization step size and motion vector information include video data. The inverse quantizer 12 dequantizes the input data according to the designated quantization step size, and outputs the inversely quantized data to an inverse DCT (discrete cosine transform) circuit 13. The inverse DCT circuit 13 processes the dequantized data using the inverse DCT to recover the video information and supplies the recovered video information to the adder 14. [

The adder 14 adds the output of the inverse DCT circuit 13 and the output of the motion compensator 15 according to the image type I, P, or B and outputs the result, that is, the motion compensated video data, frame memory bank (16).

Then, the data read from the frame memory bank 16 is rearranged in the original frame order (as shown in Fig. 14A by the switch 16E). The rearranged data is supplied to a digital-to-analog (D / A) converter 17 that converts the data to an analog video signal to be displayed on the display device 18.

18A, the output of the ECC circuit 33 is supplied to a stream detector 50 which detects image data from the stream data read from the disc 1 and supplies image type information to the control circuit 6 . In response to this information, in a specific reproduction mode, the control circuit 6 records the data of two P-pictures continuous with the I-picture in the ring buffer 5. [

Thus, three frames corresponding to I- and two P-pictures at the beginning of each GOP are written to the ring buffer 5 at high speed, and this data is decoded by the decoder 20 at any desired timing So that it becomes possible to decode effective data in a specific reproduction mode.

For example, assume that reverse playback begins with a P-picture (P ₃ ) of the original frame sequence shown in FIG. 14A. The decoded picture needs to be displayed in the following order:

P ₃ → B ₇ → B ₆ → P ₂ → B ₅ → B ₄ → P ₁ → B ₃ →

B ₂ → P ₀ → B ₁ → B ₀ → I ₀ →. . .

However, since each P- picture is coded by inter-frame prediction as described, the picture _{I 0, P 0, P 1} , and P ₂ need to be decoded prior to decoding the P- picture (P _3). Similarly, the P-pictures P ₂ and P ₃ need to be decoded prior to decoding the B-picture B ₇ . Therefore, when reverse reproduction is performed by simply decoding each picture once, as in normal reproduction, it is necessary to use a large capacity frame memory bank 16 capable of storing as many frames as the picture constituting the GOP.

To meet this requirement, the storage capacity of the frame memory bank 16 must be increased beyond what is required in normal playback mode. In addition, the decoded data must be sequentially stored in the frame memory bank in order to transfer the picture in the proper order of reverse reproduction.

Although other techniques of reverse playback can be applied to skip the B-picture and simply run into the I- and P-pictures, there is still a need to store more frames than is required for normal playback.

For this reason, the data decoding apparatus of Fig. 18 uses the same frame memory bank as used in normal reproduction to store one I-picture and two temporarily contiguous P-pictures, i.e., And performs reverse reproduction using three memory elements. The stream detector 50 provided for this purpose records an I-picture and two consecutive P-pictures in the ring buffer 5. [ However, this allows the operation and configuration of the stream detector 50 to more complexly detect the I-picture and two consecutive P-pictures.

15A shows a packetized (multiplexed) MPEG2 system flow. When a packet of the MPEG2 video stream is defined in position D3 as shown in Fig. 15C during packetization processing, the picture coding extension of picture data (V + 2) and picture header are spread in two packets as shown in Fig. 15B .

When picture headers and picture coding extensions are spread across two video packets, it is necessary to detect two video packets in order to obtain a requirement item of picture information. Further, as shown in Fig. 15A, another packet (e.g., audio packet) may exist between the two video packets so that the detection processing is complicated, thereby complicating the configuration and operation of the stream detector 50. [

According to the MPEG2 technology, video data in a frame structure in which one picture is composed of one frame and video data in a field structure in which one picture is composed of two fields can be mixed. Since picture headers are attached to all fields, picture coding extensions of two consecutive pictures and picture headers must be read to determine the data structure of the video data.

Therefore, based on the three items of information described above, namely (1) the presence or absence of a GOP header, (2) a temporal reference TR in the picture header, and (3) picture_structure information in the picture coding extension, Frame structure or a field structure.

The differential method between the frame structure and the field structure will be described in detail later.

Figures 19A and 19B show video data in field and frame structure formats, respectively. In the field structure format, one frame of video data is composed of two fields, picture header and picture data with image coding extension attached to each. In the frame structure format, one frame of video data is composed of a picture header and one frame of picture data to which a picture coding extension is attached.

In the field structure format, the TR information values in each picture header of the picture data pair are set equal to each other. Picture_structure information in the picture coding extension is " 01 " and " 10 " for the top field and the bottom field, respectively, as shown in Fig. The picture_structure information in the picture coding extension of the frame structure is " 11 " as shown in Fig.

The format of the image data (field or frame structure) can be assured by first reading the GOP header at the GOP start position and reading the picture_structure information of the picture coding extension at the beginning of the image data.

Although the image data in the frame structure can be loaded into the ring buffer 5 (Fig. 18A) by detecting a single frame, since the paired image data must be detected before it can be appropriately loaded, It is difficult to similarly load video data in the field structure constituting one frame of the frame. As a result, TR information in each picture header is read out to indicate that the two picture data units have numerically the same TR value. When these pairs are found, they are identified and loaded as paired image data.

The paired field structure picture headers are arranged in one of two different orders, top / bottom and bottom / top. This arrangement is described with reference to FIG. A GOP header (GOP H), an I-picture of a frame structure, a B-picture of a field structure, another B-picture of a field structure, a GOP header spaced apart therefrom, and an I- , Another I-picture of the field structure, and the like are sequentially recorded.

For example, in a case where a total of three frames (one I-picture and two consecutive P-pictures) are loaded into the ring buffer 5 (Fig. 18A) - Picture is detected and identified from picture_structure information ("11" in the case of frame structure), picture_coding_type in the picture header, and COP header in the picture coding extension in the picture part of the picture data.

When access is made for playback at the location identified by random access (1) in the bitstream, the picture coding extension of the first field structure B-picture and the picture header are read. At this time, the TR indicated by " 0 " is also read. Then, the picture coding extension of the two-field structure B-picture as well as the TR indicated by " 0 " and the picture header are read. Since the TR values of the two field structure B-pictures are the same, they are detected as paired data.

When an access is made at the location identified by the random access (2) in the bitstream, the picture coding extension of the first picture and the picture header are read with TR marked " 0 ". Then, the picture coding extension of the next picture and the picture header are read together with TR indicated by " 1 ". Since the numerical values of TR do not coincide with each other, data of two field structure pictures are not detected as paired data.

When access is made at the location identified by the random access 3 in the bitstream, each of the TR numeric values in the two-picture header matches each other (TR = 1) as in the case of the random access (1) . When an image structure in a picture coding extension of " 01 " or " 10 " is detected, this is regarded as a field structure and paired data is detected.

When access is made at the location identified by the random access (3) in the bitstream, the picture coding extension of the first picture and the picture header are read with TR marked " 0 ". This image data is regarded as a field structure I-picture according to the picture_structure information in the picture coding extension and the picture coding type information in the picture header.

After successive detection of the GOP header, the picture coding extension of the next picture and the picture header are read with TR marked " 0 ". Here, since the respective numerical values of the TRs coincide with each other in two consecutive images, but the GOP header exists between the two images, these two images are not considered as a pair. TR is reset to " 0 " if a GOP exists and a GOP header is not inserted between paired pictures.

As described, the stream detector 50 detects various items of information related to the picture according to a plurality of flags of the GOP header, picture header and picture coding extension to load the picture data into the ring buffer 5 Processing is executed. However, this processing routine is very complicated, making it difficult to construct the stream detector 50. [

The manner in which the stream detector 50 detects the completion of loading is described in connection with the flow chart shown in FIG. In this flowchart, it is assumed that random access is made to the input sector recorded immediately before the I-picture so that an appropriate image can be obtained immediately in response to the random access.

In step S10, the stream detector searches for picture_start_code in the picture_header to detect the picture header of the I-picture, and in step S12 a question is asked to determine whether picture_start_code has been detected. If the question in step S12 is an affirmative answer, that is, if picture_start_code is detected, the operation proceeds to step S14. However, if the question in step S12 is a negative answer, that is, if picture_start_code is not detected, the processing in step S12 is repeated until picture_start_code is detected.

In step S14, the temporary reference is read from the detected image header and the numerical value is stored in the register as TRO.

In step S16, another search (SRCH) is performed to detect the next picture for picture_start_code in the picture header, and a question is made in step S18 to determine whether picture_start_code has been detected. If the question in step S18 is an affirmative answer, that is, if picture_start_code is detected, the operation proceeds to step S20. However, if the question in step S18 is a negative answer, then in step S18 and processing is repeated until picture_start_code is detected.

A query is made in step S20 to determine whether a GOP header has been detected in the image start code and thereby determine whether the detected image data is part of a pair. If the question in step S20 is a negative answer, that is, if no GOP header is detected, the operation proceeds to step S22. However, if this question is answered positively, that is, if a GOP header is detected, the operation proceeds to step S26 because the presence of a GOP header between image data eliminates the possibility that these image units are paired.

When the temporary reference is read from the detected picture header, this numerical value is stored in the register as TR1 as indicated by step S22, and the value of the TR stored in the registers TR0 and TR1, respectively, Proceeds to the question in step S24. If the question in step S24 is an affirmative answer, that is, if two numerical values match, the operation returns to step S16 and the above-described process is repeated in connection with steps S16 to S24. The coincidence of the two numerical values indicates that the image data pair is detected.

However, if the question in the step S24 is a negative answer, that is, if the numerical values of TR are not equal, the operation proceeds to the step S26. Here, the image header of the next image is detected, and the image coding type read out from the image header is stored in the register. Operation proceeds to the question in step S28 to determine whether the stored picture coding type represents a B-picture. If the question in step S28 is an affirmative answer, that is, if the detected image is a B-picture, the operation returns to step S16 because the B-picture is not being searched, S16) to (S28).

However, if the question in step S28 is a negative answer, that is, if the detected image is not a B-picture, then the temporary reference is read in the detected image header and the numerical value is read as indicated by step S30 It is stored in register with TR2. It can be seen that this detected image is the first P-picture appearing after the I-picture.

In step S32, another search (SRCH) for picture_start_code in the picture header is made to detect the next picture, and a question is made to determine whether picture_start_code is detected in step S34. If the question in step S34 is a positive answer, the operation proceeds to step S36. However, if the question in the step S34 is a negative answer, that is, if the picture_start_code is not detected, the processing in the step S34 is repeated until the picture_start_code is detected.

In step S36, a query is made to determine whether a GOP header has been detected during the search for picture_start_code, thereby determining whether the detected image data is part of a pair. If the question in step S36 is negative, that is, if no GOP header is detected, the operation proceeds to step S38. However, if the question in step S36 is an affirmative answer, that is, if the GOP header is detected, the operation proceeds to step S42 because the presence of the GOP header between picture units eliminates the possibility that the units of this picture are paired. .

When the temporary reference is read from the image header detected in step S38, the numerical value is stored in the register with TR3, and the operation determines whether or not a match is obtained between the numerical values of the TRs respectively stored in the registers TR2 and TR3 The process proceeds to the question in step S40. If the question in step S40 is a positive answer, that is, if the two numeric values are equal, the operation returns to step S32 and the process described above in connection with steps S32 to S40 is repeated. The coincidence of the two numerical values indicates that the image data pair is detected.

However, if the question in step S40 is a negative answer, that is, if the numerical values of TR do not match, the operation proceeds to step S42 to read the image type. Operation proceeds to the question in step S44 to determine whether the stored picture coding type represents a B-picture. If the question in step S44 is an affirmative answer, that is, if the detected image is a B-picture, the operation returns to step S32 because the B-picture is not being searched, and steps S32 ) Is repeated to detect the next image.

However, if the question in step S44 is a negative answer, that is, if the detected image is not a B-picture, the temporal reference detected in the picture header is read and the numerical value is determined as TR4 Is stored in the register. It can be seen that this detected image is the second P-picture appearing after the I-picture.

Proceeding to step S48, another search (SRCH) for picture_start_code in the picture header is made to detect the next picture, and a question is made in step S50 to determine whether picture_start_code has been detected. If the question in step S50 is a positive answer, that is, if picture_start_code is detected, the operation proceeds to step S52. However, if the question in step S50 is a negative answer, the processing in step S50 is repeated until picture_start_code is detected.

During the search for picture_start_code, a determination is made whether a GOP header has been detected, and a question is then asked in step S52 to determine if the detected image data is part of a pair. If the question in step S52 is negative, that is, if no GOP header is detected, the operation proceeds to step S54. However, if the question in step S52 is an affirmative answer, that is, if a GOP header is detected, loading of the image data into the ring buffer is completed and the processing is terminated.

When the temporal reference is read from the detected picture header and the numerical value is stored in the register with TR5, as indicated by step S54, the operation is obtained as a match between the numerical values of the TRs stored in TR4 and TR5, respectively The process proceeds to the question in step S56 to determine whether or not there is an answer. If the question in step S56 is an affirmative answer, that is, if the two numerical values match, the operation returns to step S48 and the process described above in connection with steps S48 to S56 is repeated. However, if the question in step S56 is negative, that is, if the two numerical values of TR are not equal, the loading of the image data is completed and the processing is terminated.

As such, the stream detector 50 may load the bit stream by executing the above-described processing routine to load one I-picture and two consecutive P-pictures. However, it is cumbersome to carry out this complicated processing routine as described in the verbose explanation.

It is an object of the present invention to provide a method and apparatus for coding data that overcomes the above-described disadvantages of the above-described complex operations, in order to carry out a specific reproduction, for example fast forward and fast reverse reproduction.

It is yet another object of the present invention to provide a method and apparatus for decoding data to perform a specific playback that overcomes the disadvantages of the techniques described above.

It is yet another object of the present invention to provide a recording medium used in connection with a processor control apparatus to execute a specific playback without requiring a processor control apparatus to perform the above-described complicated operations.

Various other objects, advantages, and features of the present invention will become readily apparent from the detailed description, and the novel features are pointed out with particularity in the appended claims.

According to one embodiment of the present invention, an apparatus and a method for recording coded image data on a recording medium are provided. The image data is coded using intra-frame coding and / or predictive coding to provide one I-picture and two consecutive P-pictures. Position information indicating the positions of two P-pictures related to the I-picture is generated, and an I-picture, two P-pictures, and positional information are recorded on the recording medium.

In one aspect of the present invention, the position information indicates a data length which is a byte from the I picture to the end of the first P picture and / or the end of the second P picture.

According to another embodiment of the present invention, an apparatus and a method for reproducing coded image data from a recording medium are provided. Position information indicating the positions of two P-pictures related to the I-picture is detected, and a data stream including the I-picture, the two P-pictures, and the positional information is generated. The data stream is decoded and displayed.

According to another embodiment of the present invention, position information indicating the positions of one I-picture, two P-pictures, and two P-pictures associated with an I-picture is recorded on the medium and a specific reproduction is recorded relatively There is provided a recording medium used in connection with a processor control apparatus used by a processor control apparatus to execute easily and directly.

The following detailed description, which is provided by way of example, is not intended to be limiting of the invention, and will be best understood by reference to the appended drawings.

1 is a block diagram illustrating an embodiment of a data coding apparatus of the present invention;

Figure 2 shows an example of a packetized stream coded by the data coding device of Figure 1;

3 is a layout diagram of input point information.

4 is a diagram showing the syntax of a program stream map (PSM);

5 is a diagram showing the syntax of a basic strip descriptor;

6 is a diagram showing the syntax of ip_ipp_descriptor.

7 is a diagram showing a syntax of an overall descriptor;

8A and 8B are block diagrams illustrating an embodiment of a data decoding apparatus of the present invention.

FIGS. 9A to 9C are diagrams illustrating an example of a sequence of video data as a reference for explaining a method of reading video data in a fast reverse playback mode among the data decoding apparatuses of FIGS. 8A and 8B; FIG.

FIG. 10 is a timing chart of a recording / reading operation as a reference for explaining a fast reverse reproduction mode among the data decoding apparatuses of FIGS. 8A and 8B; FIG.

11A to 11C are diagrams illustrating an example of a sequence of video data as a reference for explaining a method of reading video data in the FF (fast forward) playback mode among the data decoding devices of FIGS. 8A and 8B.

FIGS. 12A to 12C are diagrams showing another example of the order of video data as a reference for explaining a method of reading video data in a fast reverse reproduction mode among the data decoding apparatuses of FIGS. 8A and 8B; FIG.

FIG. 13 is a timing chart for explaining a method of reading video data using two frame memories in a fast reverse playback mode among the data decoding apparatuses of FIGS. 8A and 8B.

14A and 14B are diagrams schematically illustrating structures of a recording frame of an image and an original interframe predictive image in the MPEG system;

Figures 15A-15C schematically illustrate an MPEG video stream.

16 is a diagram showing the structure of an image header in the MPEG system;

17 illustrates a structure of a picture coding extension in an MPEG system;

18A and 18B are block diagrams illustrating a data decoding apparatus.

Figures 19A and 19B schematically illustrate the structure of video data in field and frame formats;

20 is a diagram illustrating the contents of the image structure;

Figure 21 is a drawing of a video stream illustrating how two video formats (fields and frames) are differentiated.

22 is a drawing of a routine executed in a stream detector to load three frames of image data (one I-picture and two P-pictures) mixed with field and frame video formats;

Description of the Related Art [0002]

1: optical disc 2: pickup

3: demodulation circuit 4: sector detection circuit

5: ring buffer 6: control circuit

9: phase locked loop (PLL) circuit 10: video code buffer

11: Inverse VLC circuit 12: Inverse quantizer

13: inverse DCT circuit 14: adder

15: Motion compensator 16: Frame memory bank

17: digital-to-analog (D / A) converter 20: video decoder

32: Demultiplexer 33: ECC circuit

34: image header detector 35: image data selection circuit

50: stream detector

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram showing an embodiment of a data coding apparatus of the present invention. The audio encoder 102 compress-encodes the input audio signal supplied to the audio input terminal, and the video encoder 101 compress-encodes the input video signal supplied to the video input terminal. The encoded audio and video signals are supplied to a multiplexer 113. The stream output from the audio encoder 102 is an MPEG2 audio stream (audio layer), and the stream output from the video encoder 101 is an MPEG2 video stream (video layer), and is shown later in FIG. 15C.

The multiplexer 113 packetizes the input MPEG2 video stream and the MPEG2 audio stream by time division multiplexing to form the system flow shown in Fig. 15A.

Although not shown, a subtitle stream may also be input to the multiplexer 113 and multiplexed with the video and audio streams. In this case, the MPEG2 system flow output from the multiplexer 113 is as shown in Fig. 15A.

An input terminal of the input point data memory circuit 133A is connected to the video encoder 101 and an input point detector 31 inputs the input point data memory circuit 133A to the input point I - data on an image generation point).

A TOC (table of contents) data generator 156 generates TOC data based on the contents of the input point data memory circuit 133A. The TOC data includes a disc name in which video and audio data are recorded, a name of each chapter recorded on the disc, a start address of each chapter on the disc, a play time of the disc, a play time of each chapter, And the like.

The packetized stream output from the multiplexer 113 is temporarily stored in the digital storage medium (DSM) 110 and supplied to the TOC subscript circuit 150. The TOC subscript circuit 150 adds the TOC data to the packetized stream and supplies it to a picture header detector / program stream map (PSM) data generator-over-writer 155.

The picture header detector / PSM data generator-overwriter 155 detects the picture header and determines the picture header of the P-picture that appears second from the beginning of the input sector to the end of the first P-picture and / And generates PSM data including information indicating a data length which is a byte to an end portion. Preferably, the PSM data includes information indicating a data length that is a byte from the start of the I-picture to the end of the P-picture first appearing and / or from the start of the I-picture to the end of the P- . The generated PSM data is recorded in the area in the input sector that is previously scheduled in the packetized stream by the multiplexer 113. A detailed description of the PSM data is provided below.

The output of the picture header detector / PSM data generator-overwriter 155 is supplied to the sector header subscript circuit 151, where the packetized stream is divided into sectors having sector headers added to each sector.

The video and audio data, including the output of the sector header subscript circuit 151, i.e., all other data added thereto, as described above, is encoded for error correction by the ECC encoder 152.

The modulator 153 then modulates the encoded data from the ECC encoder 152 using 8 to 14 modulation (EFM), and the modulated data is supplied to the cutting device 154. The cutting device 154 forms a pit in the master disk 160 in accordance with the data supplied from the modulator 153 so that the packetized stream data is recorded in the master DVD disk 160. For example, a duplicate DVD disc is produced through press-molding of the master disc 160.

In this way, the data coding apparatus of FIG. 1 encodes and packetizes the audio signal and the video signal inputted thereto using time division multiplexing to produce a packetized stream. In addition, the picture header detector / PSM data generator-over-writer 155 generates and records PSM data in the packetized stream. The packetized stream is recorded on the master DVD disk 160. [

2 shows an example of an MPEG2 system flow output from a packetized stream, for example, a picture header detector / PSM data generator- Briefly, only packetized video data and audio data are shown. The audio data is inserted into specific portions of the MPEG2 system flow to ensure that the audio is not disturbed during playback, and the video data of the I-, P-, and B-pictures are inserted into the audio data.

The input point means the top (or start) portion of the I-picture, and the sector including this input point is called the input sector. In Fig. 2, the position of this input point is the input point n, the input point n + 1,. . . And so on. Since the position at which the input point information is recorded is predetermined to precede the I-picture, a complete image can be displayed immediately after the pickup reads data from the input sector.

Audio data may exist between the input point information and the I-picture, but B-pictures can not exist between them.

Fig. 3 shows the layout of the input point information. The input point information includes a pack header, a program stream directory (PSD), a program stream map (PSM), and other packets having optional system headers.

Figure 4 shows the syntax of the PSM. The PSM includes a 24-bit packet_start_code_prefix forming a unique code, an 8-bit map_stream_id, a program_stream_info consisting of an arbitrary number of overall descriptors, a stream_type, and an elementary_stream_info including an arbitrary number of elementary stream descriptors.

5 shows the syntax of an elementary stream descriptor composed of dvd_video_descriptor and ip_ipp_descriptor if the strip is video data, dvd_audio_descriptor and ISO_639_language_descriptor when the strip is audio data, and dvd_subtitle_descriptor and ISO_639_language_descriptor if the stream is subtitle data. Other items of information are also shown in Fig.

As shown in FIG. 6, the ip_ipp_descriptor includes an 8-bit descriptor_tag indicating a descriptor of ip_ipp, a description_length of 8 bits indicating the length of the descriptor, a first byte of the current input sector, and a final bit of the P- Bit_by_st_pirst_P_pic indicating the number of bytes of the current input sector, and a 32-bit byte_to_second_P_pic indicating the number of bytes from the first byte of the current input sector to the last bit of the P-picture indicated second.

As shown in FIG. 2, bytes_to_first_P_pic and bytes_to_second_P_pic represent data lengths. It can be seen that the number of offset bytes indicated by the information of bytes_to_first_P_pic and bytes_to_second_P_pic includes intermediate B-pictures and audio packets as well as I- and P-pictures as shown in FIG.

FIG. 7 shows the syntax of the overall descriptor of FIG. The PSD included in each input sector indicates the distance to the input sector preceding the current input sector and the next input sector, and the distance to the input sector after 1 second, 3 seconds, and so on. This distance is referred to as an offset address.

8A and 8B are block diagrams of a preferred embodiment of the data decoding apparatus of the present invention. Briefly, the elements shown in Figs. 8A and 8B corresponding to the elements shown in Figs. 18A and 18B are denoted by the same reference numerals.

The optical disk 1 is rotated at a predetermined rotation rate by a spindle motor (not shown), MPEG-compressed digital data recorded on the track by projecting a laser beam from the pickup 2 to a track on the optical disk 1 And is read therefrom. The digital data is EFM-demodulated by the demodulator 3 and input to the sector detector 4. [ The output of the pickup 2 is also supplied to a phase locked loop (PLL) circuit 9 to recover the clock signal supplied to both the demodulator 3 and the sector detector 4.

As described above, the digital data is recorded on the disk 1 as a sector unit of fixed length, and the sector sync and the sector header are recorded at the beginning of each sector. The division of the sector is determined from the detection of the sector address and the sector sync from the sector header, and is supplied to the control circuit 6. [ Preferably, the control circuitry is implemented in a microprocessor to perform processor control over the illustrated device.

The demodulated output is supplied to an ECC (Error Correction) circuit 33 which performs error detection and correction via the sector detector 4. [ The ECC circuit 33 supplies the error-corrected data to the ring buffer 5 to be recorded under the control of the control circuit 6. [

The output of the ECC circuit 33 is also supplied to the PSM detector 40. In a specific reproduction mode, the PSM detector 40 detects the PSM information in the input sector in the stream data read from the disc 1, and supplies the detected PSM information to the control circuit 6. In order to ensure that the length information in the stream data from the I-picture after the input sector to the second P-picture is recorded in the ring buffer 5, the control circuit 6 controls the ip_ipp_descriptor in accordance with the information on the number of offset bytes Mode to record (or load) the I- and P-pictures to the ring buffer 5 in the mode.

A focus control circuit (not shown) and a tracking servo circuit S are connected to a pickup 2 (not shown) under the control of a system controller (not shown) in accordance with a focus error signal and a tracking error signal obtained from the information read by the pickup 2 ) And the tracking of the object.

Based on the sector address of each sector detected by the sector detector 4, the control circuit 6 designates a write address for writing an appropriate sector to the ring buffer 5 with the write pointer WP. The control circuit 6 also specifies the read address at which the data is read out from the ring buffer 5 to the read pointer RP based on the code request signal obtained in the video code buffer 10 (Fig. 8B). The data read from the position designated by the read pointer RP is supplied to the demultiplexer 32. [

The coded data recorded on the disk 1 is composed of multiplexed packetized video data, audio data, and subtitle data, and the demultiplexer 32 converts the supplied data into video data, audio data, and sub- And supplies each data to the video decoder 20 (FIG. 8B), the audio decoder (not shown), and the subtitle decoder (not shown).

As a result, the video data read out from the ring buffer 5 is stored in the video code buffer 10 of the video decoder. The stream data from the I-picture to the second continuous P-picture includes video packets and other packets as shown in Fig. In a specific playback mode, data that is not needed, i.e., video data and other packets, are removed by the demultiplexer 32. [

The data stored in the video code buffer 10 includes image type information indicating image header, image type I, P, or B, and image header detector 34 in which a temporary reference TR indicating a frame order in the GOP is detected. . The detected image type information is supplied to the image data selector 35 so that only the I- and P- images are selected in a specific reproduction mode and the selected image is supplied to the inverse VLC circuit 11. [ In the normal reproduction mode, the image data selector 35 is controlled to output all the image data without performing the preliminary selection.

The data supplied to the inverse VLC circuit 11 is processed using the inverse VLC and supplied to the inverse quantizer 12. The code request signal returns from the reverse VLC circuit to the video code buffer 10 so that new data is transferred from the video code buffer 10.

The inverse VLC circuit 11 also outputs the quantization step size to the inverse quantizer 12 and outputs the motion vector information to the motion compensator 15. [ The inverse quantizer 12 dequantizes the input data according to the designated quantization step size, and outputs the inverse quantized data to the inverse DCT circuit 13. The inverse DCT circuit 13 processes the quantized data using the inverse DCT and supplies the processed data to the adder 14. [

The adder 14 adds the output of the inverse DCT circuit 13 and the output of the motion compensator 15 according to the image type I, P, or B, and outputs the resultant motion compensated video data to the frame memory bank 16. [ .

Then, the data read from the frame memory bank 16 is rearranged in the original frame order shown in Fig. 14A by the switch 16e. The rearranged data is supplied to a digital-to-analog (D / A) converter 17 so as to be converted to an analog video signal and displayed on the display device 18.

In response to the code request signal from the video code buffer, the control circuit 6 supplies the data stored in the ring buffer 5 to the video code buffer 10. When the amount of data transferred from the video code buffer 10 to the inverse VLC circuit 11 is reduced as a result of continuous data processing of a simple image representing a small amount of data, The amount of data transferred to the memory 10 is also reduced. As a result, the amount of data stored in the ring buffer 5 can be increased, and the write pointer WP can pass through the read pointer RP so as to cause an overflow of the ring buffer 5. [

In order to prevent such a problem, the control circuit 6 calculates the current amount of data stored in the ring buffer 5 based on the address positions of the write pointer WP and the read pointer RP. When the calculated amount of data exceeds a predetermined amount, the track jump determining circuit 7 determines whether or not the ring buffer 5 can overflow and transmits a track jump command to the tracking servo circuit 8. [ In response to the track jump command, the tracking servo circuit 8 causes the pickup 2 to jump the track in accordance with the storage capacity of the ring buffer 5 so as to prevent overflow or underflow of the ring buffer 5. [ This advantageously permits continuous video reproduction with uniform image quality regardless of the complexity (or flatness) of the image recorded on the disc 1.

The data transfer rate of the video code buffer 10 in the ring buffer 5 is controlled by the ECC circuit 33 so that the video code buffer 10 can transfer the code request for data transfer to the ring buffer regardless of the track jump. Is set to be equal to or lower than the data transfer speed to the data transfer unit (5).

In the normal reproduction mode, for example, I, P and B image data I ₀ , B _-2 , B _-1 , P ₀ , B ₀ , B ₁ , ). ≪ / RTI > In this example, one GOP is composed of an image 15 frame, a P-image 4 frame, and a B-image 10 frame including one frame of an I-picture. The normal reproduction of the picture is executed by sequentially reading and decoding the data coded in the recording order shown in Fig. 14B and displaying the decoded data in the order shown in Fig. 14A.

Specifically, when decoding the I-picture I ₀ , the decoded output obtained from the inverse DCT circuit 13 is supplied directly to the frame memory bank 16. However, B- picture (B _-2) the in decoding all the decoded previously used as a reference for predictive coding a picture B- (B _-2) P- picture (not shown) and the I- picture ( I _{0 is} supplied from the frame memory bank 16 to the motion compensator 15 and a motion predicted image is generated in accordance with the motion vector information supplied from the inverse VLC circuit 11. The generated motion predicted picture is supplied to an adder 14 which adds the motion predicted picture to the output of the inverse DCT circuit 13 so that the B-picture B _-2 is decoded in the frame memory bank 16 .

B- picture (B _-1) is a B- picture is decoded in the same manner as in (B _-2) is overwritten on the image B- (B _-2) stored in one of frame memories (16a to 16c) of the frame memory bank . When decoding the P-picture (P ₀ ), the I-picture (I ₀ ) is supplied to the motion compensator (15) in the frame memory bank (16) together with the motion vector information supplied from the reverse VLC circuit. Motion compensator supplies a motion predicted picture to the P- picture (P ₀₎ for the P- image supplied from the inverse DCT circuit to decode (P ₀₎ The adder 14 adds the motion predicted picture to. The decoded P-picture P ₀ is recorded in the oldest data (which may be an I-picture or a P-picture) stored in the frame memory bank 16.

As such, the pictures are sequentially decoded as described above, but are read out in the frame memory bank 16 in the original order and displayed on the display device 18 in the order shown in Fig. 14A.

In the quick playback mode, data recorded on the disc 1 needs to be decoded and displayed in the reverse order. For example, when the B-picture B ₀₇ is decoded from the disc 1 in which the video data is recorded in the order shown in Figs. 9A and 9B (only the video data is briefly shown), the desired B-picture by to decode the B ₀₇₎ since it is necessary to be a P- image has to decode the B- image ₍₀₇ B) needed to decode the P- picture (P ₀₈ and P ₀₅₎ above. However, the decoded P-picture P ₀₅ is required to decode the P-picture P ₀₈ and the decoded I-picture I ₀₂ is required to decode the P-picture P ₀₅ have. As a result, the decoding must start from the I-picture located at the beginning of the GOP. When the decoding of one GOP is ended, the operation needs to jump back to the previous GOP to continue the decoding process.

However, when reverse reproduction is performed with such a decoding technique, there is an excessive time delay in displaying an image, thereby displaying an unnatural image. The present invention solves this problem by decoding only the I- and P-pictures in the normal playback mode, but performs reverse playback using only the same three frame memories 16a, 16b, and 16c required for normal playback do. Thus, the present invention advantageously decodes a total of three pictures, i-picture and two consecutive P-pictures appearing after the input sector, instead of decoding the overall sequence of I- and P-pictures, So that fast reverse playback can be performed.

The present invention can perform fast reverse reproduction by the PSM detector 40 of a simple circuit configuration without using a stream detector having a complicated configuration as described above. Since the PSM detector 40 detects the information indicating the number of offset bytes given to the ip_ipp_descriptor in the PSM, the PSM detector 40 detects the minimum of the data stream from the I-picture immediately after the input sector to two consecutive P-pictures appearing after the I- It is possible to record only the required range in the ring buffer 5.

The operations performed by the data decoding apparatus of Figures 8A and 8B in the fast reverse (FR) playback mode are described with reference to Figures 9A-9C.

9A and 9B show the order of the video data recorded on the disc 1. In Fig. The video data shown in this figure corresponds to four GOPs and in the FR playback mode the control circuit 6 reads the video data from the disc 1 in the order in which the pickup 2 is indicated by the arrow under the video data To perform the control operation. In particular, the pick-up (2) is continuously I- picture (I _32), B- picture (B _30), B- picture (B _31), P- picture (P _35), B- image ₍₃₃ B), B Picture B ₃₄ and the P-picture P ₃₈ in this order and jumps to the immediately preceding GOP and the video data of the I-picture I ₂₂ to the second P- (P ₂₈ ). Subsequently the pickup 2 is read out to again jump to another preceding GOP, and the video image data P- (P ₁₈₎ of the second of the I- picture (I _12). Next, the pickup 2 jumps to another preceding GOP, and the video data of the I-picture I ₀₂ is read to the second P-picture P ₀₈ . Then, similarly to the above, the pickup 2 reads the video data of the I-picture located at the beginning of the next preceding GOP up to the P-picture of Fig. 2 located after the I-picture.

This operation is possible because the above-described input point information is recorded in the input sector formed at the upper end of each GOP, and the PSM detector 40 detects the ip_ipp_descriptor in the input sector and supplies the detected ip_ipp_descriptor to the control circuit 6 Do. As a result, the control circuit 6 controls the pickup 2 to read the data corresponding to the number of bytes indicated by the information of the bytes_to_second_P_pic given in the ip_ipp_descriptor from the top (or starting portion) of the input sector, 2) can read the video data instantaneously indicated by the arrows in Figs. 9A to 9C.

To access the top of the immediately executing GOP, the distance (data length) information indicated by the offset address in the PSD in the input sector is used, which means the distance to the preceding input sector.

The read video data in the range from the I-picture located at the beginning of the GOP to the second P-picture located after the I-picture is demultiplexed by the demultiplexer 32 from the audio data and other data and stored in the video code buffer 10 ). The B-picture is removed by decoding only the I- and P-pictures recorded in the frame memory bank 16 using the detection information obtained in the picture header detector 34. [ The video data thus recorded is read out from the frame memory 16 and displayed on the display device 18 in the reverse picture display order shown in Fig. 9C.

The data read / write timing for the frame memory bank 16 in the FR playback mode is described with reference to FIG. Preferably, in the frame memory bank 16, as shown in FIG. Three frame memories 16a, 16b, and 16c are provided. A frame memory (16a) of the beginning GOP (see Fig. 9B) at the end of decoding the recording of the I- picture (I ₃₂₎ is the time point (t0) and start, one frame after the time point (t1) at the top of the do. Then, in the frame memory (16b), I- picture (I ₃₂₎ decoded by reference, a P- picture (P ₃₅₎ of the recording and, beginning at the point (t1), in one frame after the time point (t2) And is terminated.

Further, the frame memory (16), P- picture (P ₃₅₎ of the recording, P- picture (P ₃₈₎ is decoded with reference to start at time point (t2), one frame after the time point (t3) Lt; / RTI > Assuming that this is the field of P- picture (P ₃₈₎ has already been recorded in the frame memory (16c) by the reading start time point, the read image of P- (P ₃₈₎ from the frame memory (16c) is the starting point (t2) and (t3). Therefore, it is possible to simultaneously read and write from the same frame memory by delaying the read timing from the write timing by one field.

P- frames read out of the memory image (P ₃₈₎ from (16c) is terminated at an intermediate point between the two time points t3 and t4, the decoded I- picture of the preceding GOP in the frame memory (16c) (I ₂₂₎ Is started at time point t3. This recording ends at time point t4 after one frame has elapsed.

Other image data can be recorded in the frame memory 16c while simultaneously reading the previously recorded image data as described above because the write timing is preferably one field delay with respect to the read timing.

Then, the image data decoded as shown in Figures 9A and 9B _{(I 32), (P 35} ), (P 38), (I 22), (P 25), (P 28), (I 12 ), (P ₁₅ ), (P ₁₈ ), (I ₀₂ ), (P ₀₅ ) _,. . And the like are recorded in the frame memories 16a, 16b, and 16c in this order. On the other hand, the image data is _{(P 38), (P 35} ), (I 32), (P 28), (P 25), (I 22), (P 18), (P 15), (I 12), . . (16b), and (16c) in order from the oldest (largest) picture number to the newer (smaller) picture number, such as, for example,

As a result, fast reverse playback is performed, and the images are displayed in the order shown in Fig. 9C. For example, three pictures per GOP can be played back in reverse when three frame memories are used.

In the reverse reproduction mode, the identification number assigned to the image is detected, and the image is read from the frame memory bank 16 in the order of the oldest (the largest number to the smallest) number. The temporary reference TR, which means a number representing the display order of the picture, is reset to each upper part of the GOP, and the value of this temporary reference is in the range of 0 to 1023.

Referring now to Figures 11A-11C, the description provides fast forward (FF) playback that is implemented in the data decoding apparatus of Figures 8A and 8B. 11A and 11B show video data of four GOPs in the order recorded on the disc 1, and arrows below the video data indicate the order of video data to be read out in the FF playback code.

In the FF playback mode, the PSM detector 40 detects the ip_ipp_descriptor in the input sector recorded at the top of each GOP, and supplies the detected ip_ipp_descriptor to the control circuit 6 as in the FR playback mode described above. The control circuit 6 controls the pickup 2 to read the data corresponding to the number of bytes indicated by the information of bytes_to_second_P_pic in the ip_ipp_descriptor from the top of the input sector so that the video data is recorded in the order indicated by the arrow in FIG. .

The B-picture given identification in each picture header is removed from the read video data and only the I- and P-pictures are decoded. The decoded I- and P-pictures are read out from the frame memory bank 16 in decoding order and read as I ₀₂ , ₀₅ , ₀₈ , ₁₂ , _{P 15), (P 18)} , (I 22), (P 25), (P 28), (I 32), (P 35), ( is displayed on the display device 18 in the order of P ₃₈₎ .

Although three frame memories are preferably included in the frame memory bank 16, the number of frame memories is not limited to these three, and any desired number can be selected. Fast forward (FF) playback is performed with I- and P-pictures of the same number as the frame memory.

FR playback executed with only two frame memories 16a and 16b in the frame memory bank 16 will be described with reference to Figs. 12A to 12C and Fig.

12A and 12B show the order in which the video data is recorded on the disc 1. Fig. The video data in Figs. 12A and 12B corresponds to 4 GOP, and in the FR reproduction mode, the control circuit 6 causes the pickup 2 to read the video data on the disc 1 in the order indicated by arrows below the video data . Particularly, the pickup 2 successively reads the video data of the I-picture I ₃₂ , the B-picture B ₃₀ , the B-picture B ₃₁ and the P-picture P ₃₅ in this order , jumps to an immediately preceding GOP, and also read out to the I- picture (I ₂₂₎ the video data P- picture (P ₂₅₎ of a first of the. Then, the pickup 2 jumps to another preceding GOP and reads the video data of the I-picture I _{12 up} to the first P-picture P ₁₅ . Next, the pickup 2 jumps to another GOP and reads the video data of the I-picture I ₀₂ to the first P-picture P ₀₅ . Then, similarly to the above, the pickup 2 reads from the video data of the I-picture located at the beginning of the next preceding GOP to the first P-picture located after the I-picture.

This operation is performed because the PSM detector 40 detects the information of bytes_to_first_P_pic in the ip_ipp_descriptor in the input sector recorded on the top of each GOP and supplies the detected information to the control circuit 6. [ Particularly, the control circuit 6 controls the pickup 2 to read the data corresponding to the number of bytes indicated by the information of the bytes_to_first_P_pic given in the ip_ipp_descriptor from the top of the input sector, 9B to read the video data in the order indicated by the arrows.

The read video data in the range from the I-picture located at the beginning of the GOP to the first P-picture located after the I-picture is demultiplexed from the audio data and other data by the demultiplexer 32, 10). Then, the B-picture whose identification is given to each picture header is removed and only the I- and P-pictures are decoded and written to the frame memory bank 16. [ The video data thus recorded is read out from the frame memory 16 in the image display order shown in Fig. 12C and displayed on the display device 18. [

Fig. 13 shows the data read / write timing for the frame memory bank 16 of two frame capacities. In Fig. 12B, the recording of the I-picture I ₃₂ decoded at the beginning of the latest GOP in the frame memory 16a starts at the time point t0 and ends at the time point t1 after the lapse of one frame. Then, the frame memory (16b) for, I- picture (I ₃₂₎ decoded by reference, a P- picture (P ₃₅₎ is, and the time point (t2) after one frame has passed from the starting point (t1) for recording Lt; / RTI >

A P- picture fields assuming that has already been recorded in the frame memory (16b) to the read start time point, is for reading a P- picture (P ₃₅₎ from the frame memory (16b) that two-hour period (P ₃₅₎ ( It starts at the midpoint between t1 and t2. Therefore, it is possible to simultaneously read and write from the same frame memory 16b by delaying the read timing with respect to the write timing by one field. In this way, other image data can be recorded in the frame memory 16b while simultaneously reading the previously recorded image data.

P- frame image from the memory (16b) (P ₃₅₎ has two reading time point (t2) and (t3) and ends at the midpoint between the frame decoded I- picture of a GOP that is prior to the memory (16b) of records (I ₂₂₎ begins at a time point (t2). This recording ends at time point t3 after one frame has elapsed.

In the I- picture (I ₃₂₎ has two time points (t2) and (t3) is read out from the midpoint of the frame memory (16a) in between, and then about one field delay from this reading start point, the frame memory (16a) The recording of the decoded P-picture P ₂₅ starts. One frame of the I- picture (I ₃₂₎ is completely read out, and then the frame image of P- (P ₂₅₎ is completely read out from the frame memory (16a) at an intermediate point between the time point (t3) and (t4) . Further, in the GOP preceding the frame of the I- picture (I ₁₂₎ is written into the frame memory (16a) between the time point (t4) and (t5).

FIG. 12A to the image data decoded as shown in Figure 12B _{(I 32), (P 35} ), (I 22), (P 25), (I 12), (P 15), (I 02), (P ₀₅ ) _,. . In the order of., And written into the frame memory (16a) and _{(16b), (P 35)} , (I 32), (P 25), (I 22), (P 15), (I 12), (P 05 ), (I ₀₂ ) _,. . (Smaller) picture numbers, such as old (largest) picture numbers, such as.

As described above, one I-picture and two or one P-picture per GOP are displayed in a specific reproduction mode. It can be seen that the present invention can be modified to decode and display only one I-picture per GOP and to eliminate both P-pictures and B-pictures. In this case, information for detecting the number of bytes to the end of the I-picture is recorded in the PSD (Program Stream Directory). In particular, in the program stream directory defined in accordance with the MPEG system (ISO 13818-1), information about the I-picture immediately after the PSD is written to the reference access unit, and PES_header_position_offset, reference_offset, and bytes_to_read Values are added together to determine the data length (total number of bytes) from the first byte of the PSD to the end of the I-picture.

When the storage capacity of the frame memory exceeds three frames, three or more frames per GOP can be decoded and reproduced in a specific reproduction mode. In this case, information indicating the data length may be recorded in the PSM and three or more P-pictures appearing after the I-picture may be accessed.

While the above example shows a jump to an adjacent GOP in a specific playback mode, a jump to a far GOP may be performed when performing a specific playback.

In the present invention in which the pickup 2 jumps in a specific reproduction mode, the video data in the video strips differ depending on the image type (I, P, or B) and the characteristic (flatness or complexity) . As a result, the search time is not fixed and difficulties may arise when FF / FR reproduction of back drainage speed is executed. In order to avoid this difficulty, the search time or display interval is measured by the system controller, and the following search is changed according to the measured time, thereby controlling the speed through feedback control. For example, if a longer time is taken in one search, the pickup 2 jumps to a position slightly farther from the unit of the GOP to obtain the required distance.

While the present invention has been described in connection with optical recording media, it will be appreciated that the recording / reproducing method and apparatus of the present invention can be used with other recording media carrying compressed video data.

Since certain regeneration, such as reverse regeneration, can be performed by the present invention with circuitry with less cost and complexity, some components of the device and the device comprising the particular circuitry can be reduced in size and consequently reduce power consumption, The structure required for heat generation and heat generation can be minimized.

In a specific playback mode, only the I-picture can be recovered, the I-picture and the P-picture can be recovered, or the I-picture and the two P-pictures can be recovered. Such a structure can be selectively exchanged so that a specific playback speed can be controlled by changing the number of images to be loaded and displayed in the frame memory. By using one or two P-pictures in addition to the I-picture, the screen can be displayed smoothly to provide a satisfactory visual presentation.

While the invention has been particularly shown and described with reference to a preferred embodiment, it will be apparent that various changes may be made therein without departing from the spirit and scope of the invention. It is intended that the appended claims be construed to include the above-described embodiments, various other ways of being described, and the like.

Claims (31)

A method for recording image data on a recording medium, Receiving the image data; Using in-frame coding and predictive coding to provide a data stream comprising at least one intra-frame coded image data (I-picture) and at least two predictive coded image data (P-pictures) in a predetermined order Coding the image data, the at least one I-picture preceding the at least two P-pictures; coding the image data; Generating position information representing positions of the at least one I-picture and the at least two P-pictures for the at least one I-picture in the data stream; and And recording the at least one I-picture, the at least two P-pictures, and the positional information on the recording medium. The method of claim 1, wherein generating the position information comprises: generating the position information from the at least one I-picture to a first end of the at least two P-pictures and from the at least two P- And generating positional information indicating a data length up to a second end of the image. 3. The method of claim 2, wherein the recording step writes the at least one I-picture and the at least two P-pictures to a sector, and wherein generating the position information comprises storing the at least one I- Generating position information indicating a data length from the beginning of the sector to the end of the first P-picture and from the beginning of the sector to the end of the second P-picture, And the image data is recorded on the recording medium. 4. The method of claim 3, wherein the recording step comprises recording the position information in the sector. 5. The method of claim 4, wherein the coding step comprises bi-directional predictive coding of image data to provide a plurality of bidirectionally predictively coded pictures (B-pictures) in the data stream . 5. The method of claim 4, wherein the at least one I-picture and the at least two P-pictures each record image data on a recording medium constituting a frame. An apparatus for recording image data on a recording medium, Means for receiving the image data, Using in-frame coding and predictive coding to provide a data stream comprising at least one intra-frame coded image data (I-picture) and at least two predictive coded picture data (P-pictures) Means for coding the image data, the at least one I-picture preceding the at least two P-pictures; means for coding the image data; Means for generating position information indicative of positions of the at least one I-picture and the at least two P-pictures for the at least one I-picture in the data stream; And means for recording the at least one I-picture, the at least two P-pictures and the positional information on the recording medium. 8. The method of claim 7, wherein the generating means is configured to generate the at least one P-picture from the at least one I-picture to a first end of the at least two P- And to generate positional information indicative of a data length up to the end of the recording medium. 9. The apparatus of claim 8, wherein the recording means is operable to record the at least one I-picture and the at least two P-pictures in a sector, and the generating means is operable to write the at least one I- Image that is operable to generate positional information indicative of a data length from a beginning of a sector to an end of the first P-picture and from a beginning of the sector to an end of the second P- An apparatus for recording image data on a medium. 10. The apparatus of claim 9, wherein the recording means is operable to record the position information in the sector. 11. A recording medium as claimed in claim 10, wherein the coding means is operable to record in a recording medium operable to code image data using bidirectional predictive coding of image data to provide a plurality of bidirectionally predictively coded pictures (B-pictures) An apparatus for recording image data on a recording medium. 11. The apparatus of claim 10, wherein the at least one I-picture and the at least two P-pictures each record image data on a recording medium constituting a frame. A method of reproducing image data from a recording medium, the image data being recorded in a data stream representing a group of pictures (GOP), each GOP including at least one intra-frame coded image data (I- (P-pictures), said at least one I-picture preceding said at least two P-pictures, said method comprising the steps of: Reproducing the data stream; Detecting position information included in the data stream and indicating position of the at least one P-pictures and the at least one I-picture for the at least one I-picture in the data stream; Detecting the at least two P-pictures and the at least one I-picture from the data stream and deriving coded image data therefrom in response to the location information; Decoding the coded image data to provide decoded image data, and And displaying the decoded image data on the recording medium. 14. The method of claim 13, wherein said detecting step comprises: detecting from the at least one I-picture to a first end of the at least two P-pictures and from at least one I-picture to a second end of at least two P- And detecting position information indicating a data length from the recording medium. 15. The method of claim 14, wherein: the reproducing step is to reproduce the data stream from a sector, and the detecting step includes detecting a portion of the data stream from the beginning of the sector including the at least one I- And detecting position information indicating a data length from a start of the sector to an end of the second P-picture. 16. The method according to claim 15, wherein the position information is recorded in the sector. An apparatus for reproducing image data from a recording medium, the image data comprising a plurality of GOPs each comprising at least one coded image data (I-picture) and at least two predictive coded image data (P-pictures) The apparatus comprising: means for recording a data stream representing a group of pictures (GOP) Means for reproducing the data stream, Means for detecting positional information contained in the data stream and indicating position of the at least two P-pictures with respect to the at least one I-picture in the data stream; Means for detecting the at least one I-picture and the at least two P-pictures in response to the positional information from the data stream and for deriving coded picture data therefrom; Means for decoding the coded image data to provide decoded image data, and And means for displaying the decoded image data. 18. The apparatus of claim 17, wherein the detecting means is arranged to detect the at least one P-picture in the at least one I-picture and the first end of the at least two P- The position information indicating the length of the data up to the end of the second portion of the image data. 19. The apparatus of claim 18, wherein the reproducing means is operable to reproduce the data stream from a sector, and wherein the detecting means is operable to detect, at a beginning of the sector comprising the at least one I- And detecting positional information indicative of a data length from the beginning of the sector to the end of the second P-picture at the beginning of the sector. 20. The apparatus according to claim 19, wherein the position information is recorded in the sector. 21. The apparatus of claim 20, wherein the coded image data in the reproduced data stream includes a plurality of bidirectionally predictively coded image data (B-pictures). 21. The apparatus of claim 20, wherein each of the at least one I-picture and the at least two P-pictures reproduce image data from a recording medium constituting a frame. An apparatus for recording and reproducing image data on a recording medium, Means for receiving the image data, In-frame coding and / or predictive coding to provide a data stream comprising at least one intra-frame coded image data (I-picture) and at least two predictive coded image data (P-pictures) Means for coding the image data, wherein the at least one I-picture precedes the at least two P-pictures; Means for generating positional information indicating positions of the at least one I-picture and the at least two P-pictures for the at least one I-picture in the data stream; Means for recording the at least one I-picture, the at least two P-pictures, and the location information on a recording medium; Means for reproducing the data stream from the recording medium, Means for detecting the location information from the reproduced data stream, Means for detecting the at least one I-picture and the at least two P-pictures from the data stream and deriving coded picture data therefrom in response to the detected position information; Means for decoding the coded image data to provide decoded image data, and And means for displaying the decoded image data on the recording medium. 24. The apparatus of claim 23, wherein the means for generating location information comprises means for generating at least one P-picture from the at least one I-picture to a first end of at least two P- And to generate positional information indicative of a data length up to a second end of the images. 25. The apparatus of claim 24, wherein the recording means is operable to write the at least one I-picture and the at least two P-pictures in a sector, and the generating means is operable to write the at least one I- Generating a positional information indicative of a data length from the beginning of the sector to the end of the first P-picture and from the beginning of the sector to the end of the second P-picture, And records the image data on the recording medium. 26. The apparatus of claim 25, wherein the recording means is operable to record the position information in the sector. 27. The computer-readable medium of claim 26, wherein the coding means is operable to code image data using bidirectionally predictive coding of image data to provide a plurality of bidirectionally predictively coded pictures (B-pictures) in the data stream And recording and reproducing the image data on the recording medium. 27. The apparatus of claim 26, wherein the at least one I-picture and the at least two P-pictures each record and reproduce image data on a recording medium constituting a frame. 24. The apparatus of claim 23, wherein the detecting means is arranged to detect at least one of the at least two P-pictures in the at least one I-picture and the first end of the at least two P- And to detect positional information indicating a length of data up to the end of the second recording and reproducing the image data on the recording medium. 30. The apparatus of claim 29, wherein the reproducing means is operable to reproduce the data stream from a sector, and wherein the position information detecting means is operable to detect the position of the first P < RTI ID = 0.0 > - record and reproduce image data on a recording medium which is further operable to detect position information indicating the length of the data from the beginning of the sector to the end of the picture and from the beginning of the sector to the end of the second P- Device. 31. The apparatus according to claim 30, wherein the position information is recorded in the sector and the image data is recorded on the recording medium.